Background: In our previous study, a marked association between Helicobacter pylori seropositivity and functional polymorphisms of the secretor and Lewis genes (odds ratio [OR], 0.32; 95% confidence interval [CI], 0.14-0.70 for se/se genotype relative to Se/Se genotype; and OR, 10.3; 95% CI, 3.16-33.8 for high-risk group defined by the combination of Se and Le relative to low-risk group) had been observed for 239 non-cancer Japanese outpatients of the gastroenterology clinic (OGC) undergoing gastroscopy at Aichi Cancer Center Hospital.
Methods: The present study was a confirmatory study to examine the association for 679 first-visit outpatients (FVO) of all clinics at the same cancer hospital and for 465 health checkup examinees (HCE) in the same city.
Results: The associations between H. pylori seropositivity and the Se and Le genotypes were nonsignificant or even in the opposite direction among the FVO (OR, 1.52; 95% CI, 1.00-2.32 for se/se genotype relative to Se/Se genotype; and OR, 0.77; 95% CI, 0.43-1.40 for the high-risk group defined similarly to the previous study), and among the HCE (OR, 1.25; 95% CI, 0.75-2.07; and OR, 1.07; 95% CI, 0.50-2.26, respectively). The discrepancy between the previous and present results was not explained by the difference in the distributions of age, sex, smoking, and H. pylori seroprevalence.
Conclusion: Even in the same ethnic group, different sources of subjects may demonstrate inconsistent findings due to an unidentified effect modification. Inconsistent findings have rarely been reported by the same research group, but they are very important to understand the whole picture of the association under study.